[Research Paper] 대한금속・재료학회지 (Korean J. Met. Mater.), Vol. 55, No. 3 (2017), pp.202~208 DOI: 10.3365/KJMM.2017.55.3.202 202 Simulation and Experimental Study on the Steady Conduction Solution for Continuous Rheo-Casting for A356 Alloy Do Minh Duc, Nguyen Hong Hai, and Pham Quang * Key Laboratory of Metal Materials Technology, Hanoi University of Science and Technology (HUST), No.1 Dai Co Viet, Hanoi, Vietnam Abstract: Computational fluid dynamic modeling of a continuous rheo-casting technology was conducted, consistent with the manufacturing of 3 -thin plates made of aluminum alloy A356. The A numerical simulation on of the stabilizing time of the material crystallization was carried out using the ANSYS FLUENT code. Solidification and melting models were simulated with heat transfer and solid-liquid phase transformation involving the latent heat of crystallization were simulated. The calculated temperature distribution and the evolution of cooling rate through the material were examined and used to clarify their influence on microstructure, and further investigated with hardness testing. The thickness of the mushy zone was determined for the steady conduction solution of the continuous rheo-casting process. † (Received December 29, 2015; Accepted August 9, 2016) Keywords: semisolid processing, solidification, solid - liquid phase transition, computer simulation 1. INTRODUCTION Semi-solid metal (SSM) processing has emerged as an attractive method for near-net-shape manufacturing due to its distinct advantages over conventional near-net-shape forming technologies. These advantages include lower cycle time, increased die life, reduced porosity, limited solidification shrinkage, improved mechanical properties, etc. The SSM processing techniques not only allow the production not only of complex dimensional details (e.g., thin-walled sections) usually associated with conventional high-pressure die castings, but also of high integrity castings that are currently attainable only with squeeze and low-pressure permanent mold casting. There are two primary semi-solid processing routes: thixocasting and rheocasting. Continuing efforts to reduce processing costs have led to the development of several rheocasting (also termed slurry-on-demand) processes. These include UBE’s New Rheocasting (NRC) [1], Idra-Prince’s Semi-Solid Rheocasting (SSR) [2], THT Presses’ Sub-Liquidus Casting (SLC TM ) [3], and Alcan’s Swirl Enthalpy Equilibration Device (SEED) [4], *Corresponding Author: Pham Quang [Tel: +84-4-38692-033, E-mail: [email protected]] Copyright ⓒ The Korean Institute of Metals and Materials as well as the Continuous Rheoconversion Process (CRP TM ) [5], developed by ACRC/MPI. In the CRP TM is a process, where the molten metal flows through a reactor prior to casting. The role of the reactor is to ensure that a extensive nucleation takes place and that the nuclei are well distributed throughout the system, prior to entering the casting cavity. The CRP TM has been successfully applied in to hyper-eutectic Al-Si alloys (i.e. 390 alloy) where two liquids of equal or different compositions and temperatures are mixed in the reactor and create a SSM slurry. In 1967, Mizikar [6] introduced an approach that about involved the effect of convection currents in the heat transfer model, and a heat transfer method known as effective thermal conductivity. In this method, instead of calculating fluid flow by with the Navier-Stokes equation, an increase the in heat transfer fluid is associated with the changing thermal conductivity of the steels. The thermal conductivity between the liquid metal and the mushy zone is raised by a quantity , and the heat transfer process does not occur when the A value is 1. This method of thermal conductivity is described with Equation 1 [7,8], (1)
7
Embed
Simulation and Experimental Study on the Steady Conduction ...kjmm.org/upload/pdf/KJMM-2017-55-3-202.pdf · Simulation and Experimental Study on the Steady Conduction Solution for
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Simulation and Experimental Study on the Steady Conduction Solution for Continuous Rheo-Casting for A356 Alloy
Do Minh Duc, Nguyen Hong Hai, and Pham Quang*Key Laboratory of Metal Materials Technology, Hanoi University of Science and Technology (HUST), No.1 Dai Co Viet,
Hanoi, Vietnam
Abstract: Computational fluid dynamic modeling of a continuous rheo-casting technology was conducted, consistent with the manufacturing of 3 -thin plates made of aluminum alloy A356. The A numerical simulation on of the stabilizing time of the material crystallization was carried out using the ANSYS FLUENT code. Solidification and melting models were simulated with heat transfer and solid-liquid phase transformation involving the latent heat of crystallization were simulated. The calculated temperature distribution and the evolution of cooling rate through the material were examined and used to clarify their influence on microstructure, and further investigated with hardness testing. The thickness of the mushy zone was determined for the steady conduction solution of the continuous rheo-casting process.
†(Received December 29, 2015; Accepted August 9, 2016)
Heat transfer coef.: casting-roller [W.m-2.K-1] 2000Contact resistance: casting-roller [m2.-K.W-1] 5
Heat transfer coef.: container-environment [W.m-2.K-1] 100Heat transfer coef.: roller-environment [W.m-2.K-1] 50
Fig. 2. Simulation results of (a) temperature field and (b) mass (liquid) fraction
between the melt with the container and between the melt
with the roller were determined through by experiment.
2.4 Microstructure Characterisation and Hardness
Testing
To establish determine any relationship between the
microstructural features and properties of the alloy, samples
were cut from identical test positions. After polishing and
etching with 10% NaOH, a Zeiss Axiovert 25 microscope
was used to observe the microstructure and determine grain
size. The as-cast samples were kept at room temperature for
some weeks before conducting hardness measurements using
a Vickers tester with a load of 0.1 kg. The Vickers hardness
values were the average of at least five indentations.
3. RESULTS AND DISCUSSION
As shown in Figures. 2a, at 0.1, 1, 2, 2.9, 3.0 and 3.1 s, the
temperature contours are fairly uniform through the melt front
and solid material. In the continuous rheo-casting process, it
is important to pull out the solidified metal at the proper time.
If the metal is pulled out too early, it may still be in a mushy
state. If it is pulled out too late, it could solidify in the casting
pool and cannot be pulled out as a whole. An optimal pulling
rate can be determined from the profile for liquidus and
solidus temperatures.
The entering rate of entry of the liquid metal at from the
left of the heat exchange domain must be balanced with the
rate of pulling rate of the solidified metal from the right and
this can be ensured by appropriate pulling rate. After 3
seconds, the equilibrium position of the melt front is well
established, as shown in Figs. 2b.
The melt stream, characterized by the liquid fraction,
becomes more narrow from the beginning to the end of the
cooling channel. However, the solid stream, identified by the
solid fraction and grain size, is behaves the opposite. As
대한금속・재료학회지 제55권 제3호 (2017년 3월) 206
Fig. 3. (a) Cooling curve in cross section (points 1 to 5) and optical micrographs of samples at (b) the contact surface, (c) the middle and (d) the center of casting
Fig. 4. Display of temperature contours to determine thickness of mushy zone (a),profile of density (b) and corresponding hardness values (c)
discussed above, nucleation occurs in the input position by
due to sudden chilling and then the nuclei are pushed by the
melt towards the end of the cooling channel, and grow to
grains afterwards.
Figure. 3a represents the temperature profile resulted
measured at the points 1, 2, 3, 4 and 5 in the cooling channel.
The largest cooling rate, of about 1050 K/s, was reached at
the contact surface between the casting and roller (point 5)
about 1050 K/s; for the other positions 1, 2, 3 and 4, the
cooling is slower, with a reduced cooling rate in the range of
(110 ÷ to 115) K/s.
In the stability studies involving, steady conduction state,
the a crystallization process commonly occurs immediately as
the liquid metal is poured into the container because the heat
exchange immediately begins between the metal and the
container. The liquid metal temperature comes down
gradually to the solidification temperature and crystallization
occurs from the contact surfaces with the cooling roller
toward the casting center.
Microstructures of the casting from the periphery to its
center (point 5 to point 1) are shown in Fig. 3. The grain size
at the contact surface (point 5) is finer than that at the casting
center (point 1), but several traces of dendrites are still visible
on the surface (Fig. 3b). The best globular grains are present
in the middle (points 2-4, Fig. 3c). In the casting center the
grains became rounder but coarse (Fig. 3d).
This difference in microstructures is explained by the
change of in cooling rate from between the surface area in
contact with the cooling roller and the casting center. Because
the solidification proceeds from the contact surface to the
casting center, the cooling rate of the first must be always
higher than that of the second. Moreover, a dendrite
fragmentation is expected to take place, creating the ideal
nuclei and leading to the globular microstructures, which are
well developed at in the middle section, as shown in Fig. 3c.
In the continuous rheo-casting process, the solidified metal
is continuously pulled out from the solidified domain. The
liquid pool, the mushy zone and the solidified shell are shown
in Figure. 4 (a). Consequently, the solid metal will have a
finite velocity that which needs to be considered taken into
account by in the enthalpy-porosity technique.
As mentioned, the enthalpy-porosity approach treats the
solid-liquid mushy zone as a porous medium with the a
porosity that is equal to the liquid fraction. A suitable sink
term is added in the momentum equation to consider reflect
the pressure drop due to the porous structure of the mushy
zone. For continuous rheo-casting applications, the relative
velocity between the molten liquid and the solid is used in the
momentum sink term, rather than the absolute velocity of the
liquid.
The mushy zone constant Amush measures the amplitude of
207 Do M inh Duc, Nguyen Hong Hai and Pham Quang
the damping; the higher this value, the steeper the transition
of the metal velocity to zero as it solidifies. Very large Amush
values may cause the liquid metal to oscillate. The pull
velocity is included to the account for the movement of the
solidified metal as it is continuously withdrawn in the
continuous rheo-casting process. The presence of this term in
the equation allows the newly solidified metal to move with
the pull velocity. The momentum sink due to the reduced
porosity in the mushy zone takes the following form [25]:
, (10)
where is a small number (0.001) to prevent division by
zero; Amush is the mushy zone constant, and is the solid
velocity which corresponds to the pulling of the solidified
material out of the solidified domain (also referred to as the
pull velocity).
An exact calculation of the pull velocity for of the solid
metal is dependent on the Young’s modulus and Poisson’s
ratio of the solid and the forces acting on it. ANSYS
FLUENT uses a Laplacian equation to approximate the pull
velocities in the solid region based on velocities at the
boundaries of the solidified region [25]:
∇ (11)
The density profiles corresponding in to different times (at
1.5, 2.0, 2.5 and 3 s) are shown in Fig. 4b. It is noted that the
density increased from the point 1 to the point 5. The
distribution of hardness test shown in Fig. 4c is in good
agreement with that of density. The measured hardness values
are 79 HV at the contact surface (point 5) and 64 HV in the
casting center (point 1). Obviously, grain size plays a
dominant role in this finding. The difference in grain size
depends on the cooling rate mentioned above (Fig. 2b) from
point 1 to point 5, i.e., from the casting center to the contact
surface with of the cooling roller. This difference is mainly
caused by the evolution of heat and temperature.
A heat transfer model with including contact resistance
between the liquid metal and the cooling roller was simulated
and was proved perfectly in this study.
This research approach and results shed light on various
advanced casting processes [29, 30] and modeling [31-34].
4. CONCLUSIONS
The ANSYS FLUENT code uses the following boundary
conditions for computing the steady conduction solution: the
liquid-solid interface, the heat transfer with including contact
resistance between the liquid metal and the cooling roller and
at a zero-gradient velocity. The solidification model was used
successfully for modeling the continuous rheo-casting
processes, where the solid metal is continuously pulled out
from the casting domain.
The steady conduction solution was simulated and verified
using experiments. The following results were obtained:
(1) The temperature field, mass (liquid) fraction and
temperature profile were simulated; the cooling rates at the
casting surface were calculated to be about 1050 K/s, and
inside the casting - to be about (110 ÷ 115) K/s; their
correlation with the microstructures was clarified.
(2) The temperature profile was displayed to determine the
thickness of the mushy zone, which formed during the time
interval from 0 to 3.5 s; the metal density changed
correspondingly from 2425 kg/cm3 at the center to 2650 at the
contact surface. The results of the hardness test were in the
full agreement with those for density.
(3) Using the constant value Amush (for 2.9 - 3.1 s) approach
and an unchanging pull velocity unchanged over the casting
domain, seems to be less expensive than using ANSYS
FLUENT to calculate the pull velocities.
ACKNOWLEDGEMENT
This study was carried out at Key laboratory of Metal
Materials Technology, the Hanoi University of Science and
Technology under sponsorship of Materials research program,
Department of Science and Technology of the Hanoi city.
대한금속・재료학회지 제55권 제3호 (2017년 3월) 208
REFERENCES
1. UBE Industries Ltd., in European Patent, EP 0 745 694 A1, 117 (1996).
2. J. A. Yurko, R. A. Martinez, and M. C. Flemings, 8th International Conference on Semi-Solid Processing of Alloys and Composites (eds. A. Alexandrou, G. Georgiou, M. Makhlouf and D. Apelian), Limassol, Cyprus, JLJ Technologies and WPI MPI, USA (2004).
3. J. L. Jorstad, 8th International Conference on Semi- Solid Processing of Alloys and Composities, (eds. A. Alexandrou, G. Georgiou, M. Makhlouf and D. Apelian), Limassol, Cyprus, JLJ Technologies and WPI MPI, USA (2004).
4. D. Doutre, J. Langlais, and S. Roy, 8th International Conference on Semi-Solid Processing of Alloys and Composites, (eds. A. Alexandrou, G. Georgiou, M. Makhlouf and D. Apelian), Limassol, Cyprus, JLJ Technologies and WPI MPI, USA (2004).
5. Q. Y. Pan, M. Findon, and D. Apelian, 8th International Conference on Semi-Solid Processing of Alloys and Composites, (eds. A. Alexandrou, G. Georgiou, M. Makhlouf and D. Apelian), Limassol, Cyprus, JLJ Technologies and WPI MPI, USA (2004).
6. E. A. Mizikar, Trans. Met. Society AIME 239, 1747 (1967).
7. V. Karlinski, S. Louhenkilpi, and J. A. Spim, Mathematical Formulas, Parameter Study and Effect of Steel Grade 40th Steelmaking Seminar International, Brazil (2009).
8. S. Louhenkilpi, Mater. Sci. Forum 414-415, 445 (2003). 9. J. E. Lait, J. K. Brimacombe and F. Weinberg, Ironmaking
& Steelmaking 2, 90 (1974).10. J. Szekely and V. Stanek, Met. Trans. 1, 119 (1970).11. S. K. Choudhary, D. Mazumdar, and A. Ghosh, ISIJ
International 33, 764 (1993).12. E. Y. Ko, J. Choi, J. Y. Park, and I. Sohn, Met. Mater.
Int. 20, 141 (2014).13. J. J. Park, Met. Mater. Int. 20, 317 (2014).14. J. H. Cho, H. W. Kim, C. Y. Lim, and S. B. Kang, Met.
Mater. Int. 20, 647 (2014).15. G. Timelli, G. Camicia, S. Ferraro, and R. Molina, Met.
Mater. Int. 20, 677 (2014).16. F. H. Yekta and S. A. Sadough Vanini, Met. Mater. Int.
21, 913 (2015).17. N. H. Hai, D. M. Duc, N. N. Tien, and N. T. Tai, Adv.
Tech. Mater. Produ. Proc., (eds. V. Y. Shchukin, G. V. Kozhevnikova and K. D. Kirdun), Belarus National Academy of Sciences, 209 (2011).
18. P. Quang, N. H. Hai, and D. M. Nghiep, Adv. Tech. Mater. Produ. Proc., (eds. V. Y. Shchukin, G. V. Kozhevnikova and K. D. Kirdun) Belarus National Academy of Sciences, 225 (2011).
19. D. M. Duc, N. H. Hai, and P. Quang, Sci. Tech. Metals 46, 37 (2013).
20. N. T. Tai, D. M. Duc, N. H. Hai, and P. Quang, J. Sci. Tech. 5A, 201 (2013).
21. N. T. Tai, D. M. Duc, N. H. Hai, and P. Quang, J. Sci. Tech. 5A, 194 (2013).
22. D. M. Duc, N. T. Tai, N. H. Hai, and P. Quang, Proc. of the 15th International symposium on Eco-materials processing and design (ISEPD2014) (eds. B. T. Long, B. V. Ga and K. Niihata) 404 (2014).
23. J. Blazek, “Computational Fluid Dynamics: Principles and applications”, Elsevier Science Ltd., Oxford, UK (2001).
25. ANSYS FLUENT 14.1, “Theory guide”, Ch. 17. Solidification and melting, ANSYS, Inc. (2011)
26. ANSYS FLUENT 14.1 Software, ANSYS, Inc. (2011)27. T. Haga, J. Mater. Pro. Tech. 130–131, 558 (2002).28. E. A. Brandes, “Smithells Metals Reference Book”
(Seventh Edition) Published by Butterworth & Co. Ltd (1983).
29. M. Azizieh, R. Bahadori, M. Abbasi, E. Y. Yoon and H. S. Kim, Inter. J. Cast Metals Research 28, 345 (2015).
30. J. H. Lee, H. S. Kim, B. Cantor and C. W. Won, Mater. Sci. Eng. 338A, 182 (2002).
31. H. S. Kim, Mater. Trans. 42, 536 (2001).32. H. S. Kim, Y. S. Lee, S. I. Hong, A. A. Tarakanova, and I.
V. Alexandrov, J. Mater. Proc. Technol. 142, 334 (2003).
33. P. Quang, Y. G. Jeong, S. C. Yoon, S. H. Hong, and H. S. Kim, J. Mater. Proc. Tech. 187-188, 318 (2007).
34. S. C. Yoon, A. V. Nagasekhar, J. H. Yoo, M. I. A. E. Aal, M. Vaseghi, and H. S. Kim, Mater. Trans. 51, 46 (2010).